Ischemic cardiomyopathy affects approximately 3 million people in the United States. This form of heart failure is the result of myocardial infarcton or severe coronary heart disease that reduces the viability and function of the heart. Ischemic cardiomyopathy is associated with poor long-term survival when patients with viable myocardium are not revascularized. By imaging myocardial blood flow and glucose metabolism and seeking flow-metabolism mismatches, positron emission tomography (PET) method has been established as the gold standard of assessing myocardial viability for selecting patients who can benefit most from surgical revascularization. Current PET method employs two separate static scans with two different radiotracers for generation of the flow-metabolism image pair. While the image of glucose metabolism is acquired using the most widely used radiotracer 18F- fluorodeoxyglucose (FDG), myocardial blood flow imaging with the radiotracer 13N-ammonia or rubidium-82 suffers from limited clinical availability. In addition, the imaging protoco of two separate imaging sessions is time consuming and resource intensive. As a result, myocardial viability via PET is currently under-utilized in clinic despite its high accuracy and th fast-growing installation of PET/CT scanners in the past decade. In this project, we propose to develop a novel PET method for myocardial viability assessment that only uses a single injection of FDG without the need of a flow- specific radiotracer. We hypothesize that myocardial blood flow can be derived from the quantitative kinetic parameters of dynamic FDG PET. We will develop a new multi-variable prediction model using statistical machine learning to predict myocardial blood flow from dynamic FDG PET data. We will also develop a shortened dynamic FDG PET protocol to improve practicality. This innovation will provide the flow-metabolism image pair for myocardial viability assessment in a clinically favorable time, cost and with reduced radiation dose. Success of this research will make PET assessment of myocardial viability more widely available in clinic with easier access, lower radiation dose, cheaper imaging cost and shorter clinical visit time as compared with conventional two-session protocols, thus improving our clinical practice of treating ischemic cardiomyopathy.